The parts nobody should have to remember, running on a schedule or on an event. Two engineers build it at your business, on your own records, and it stops and asks when a rule runs out.
The Monday report. Chasing the deposit. Reordering at the reorder point. The follow-up two days after delivery. None of these fail loudly. They simply do not happen the week that person is on leave, and nobody can say afterwards what it cost.
AI automation, as we build it, is one of those jobs taken off a person and run by an agent on the business’s own records. Data first, then connections, then the agent. Anything that reaches a customer, carries money or changes a record waits for a person to approve it.
At Irelax, stop three ran late. The remaining stops were re-planned at 09:48, total drive time came down 18 minutes, and the driver confirmed with one tap.
A five-tier win-back the sales lead approves, and a dispatch agent that re-plans the day’s Auckland run when a stop runs late.
Tyre lines reconciled into one catalogue across three branches, then picking on the phone already in the picker’s pocket.
AM-series componentry rules captured once as data. A marked-up elevation now becomes an order and cut plan in Archimax’s own item codes.
Student and enquiry records merged from messaging, email and payments into one customer record — one record behind the whole business, from the first enquiry to the after-sales visit.
Each job runs on the same record. It runs on a schedule or on an event, and nobody starts it.
Orders, addresses, promised windows and job durations sit in one record, so the day is sequenced from what is actually on the truck.
The crew carries the run on a phone. When a stop runs long, the next stops are re-planned and the change is confirmed by the office.
Delivery routing →
Each week the agent reads the customer record and finds who is due: bought five or more years ago, nothing since, no open complaint. It drafts the message for each one.
The sales lead approves before anything is sent. Customers who opted out, and anyone with an open ticket, are held back automatically.
Customer win-back →
One catalogue across branches, in the codes the counter actually says on the phone.
Before suggesting a purchase, the agent checks whether another branch already holds the stock. The order waits for approval.
Reorder before it runs out →
Every morning, payments are matched to orders by amount, date, reference and customer, with the delivery status alongside.
A deposit waiting on a delivery is not treated as a debt, and nobody who has paid gets chased.
Chase the invoices →The characteristic failure of automation is silence. It ran, it half-worked, nobody was told, and the discrepancy surfaces as a customer complaint six weeks later. An automation that fails loudly is worth more than one that is slightly more capable. Four things to ask of whoever builds yours, with what each one costs when we do it:
Someone at your business is told when it breaks. An automation nobody owns becomes a mystery within a year, and mysteries get switched off. The cost: you have to nominate a person, and they have to agree.
If a sync breaks or a rule cannot be applied, it halts and messages that person rather than doing its best guess. The cost: more interruptions in month one than you expected, while the rules are still loose.
Automating a bad process produces wrong answers faster and with more authority. Sometimes the right outcome of a fortnight is that you change how the job is done and we build nothing. The cost to us is a project that ends without software, and that has happened.
These are built as code in your own infrastructure, not as a Zapier account nobody can read. The cost: you cannot build a new one yourself on a Friday afternoon. Changes come through us, or through your upkeep hours.
Staff keep the tools they know. The record sits underneath them, and the automations read and write through the same connections.
Every engagement follows the same four steps, in the same order, whatever the size of the job.
The full terms are on How we work; the other nine offerings are on Offerings; who owns the data is on Data ownership.
Four steps, two minutes. Within two working days: can AI take them on, what it needs, roughly what it costs.